Development Of Emergency Patient Satisfaction Questionnaire In General Hospital

Objective: Patient satisfaction is an indicator for evaluating the quality of care in the emergency department. It is this perception of satisfaction that becomes the basis for future emergency department choice or the recommendation of a specific emergency department to other potential patients. Therefore, it is meaningful to develop reliable and valid Emergency Patient Satisfaction Questionnaire (EPSQ) as an instrument of measurement for evaluating the quality of medical care and meeting the urgent needs in modernlized administration of the hospitals.Subjects: The 33 emergency patients from two general hospital graded 3A of Guang Zhou and Fo Shan were surveyed by Emergency Questionnaire about items' essentiality from Nov 2003 to Jan 2004. The 33 managers of hospital in GuangDong province and 26 doctors and nurses of emergency department were surveyed by the same questionnaire meanwhile. In Mar 2004 pilot survey was conducted at NanFang hospital in GuangZhou and 38 emergency patients were interviewed. Formal survey was conducted at 4 general hospitals in Guang Zhou , Zhang JiaKou and Bao Ding from April to October 2004. And 465 emergency patients were interviewed. Methods: The study was carried out over a 11-month period .The interview was carried out on the spot .Respondents were asked to evaluate the following satisfaction factors: ambulance service, doctors' attitude and technology,nursing care,facilities,cost and medical outcome ,auxiliary,patient choice,waiting time.Most items had 5 choiced answer (5=excellent,4=very good,3=good,2=fair,1=poor) .Forrespondendents very yong or very old,a responsible adult who had accompanied them to the emergency department was asked to respond to the survey. Selection of items and factors in the questionnaire was based on comprehensive consideration with the response rate, critical rate (CR) value, variation degree,correlation and factor analysis.Internal consistency of factors and whole questionnaire was examined by calculating Cronbach alpha. Construct validity was evaluated by factor analysis as well as correlation among factor scores,EPSQ scores and scores of item ' general satisfaction'.Construct and relations of factors were confirmed with Structural Equation Model method. Demography was analyzised with CHAID analysis. Chi-square,Student's t-test, analysis of variance and Kruskal-Wallis were used to compare the respective demorgaphy. Except CHAID analyse was performed using Answertree 2.0, Structural Equation Model was performed using AMOS 4.0,All statistic analyses were conducted by the SPSS 10.0 softwareResult: The first survery was delivered to the 33 patients,the 33 managers and the 26 staffs of emergency department. The 26 items were selected to consist of the questionnaire based on the principle of statistical technique. The questionnaire was composed with 8 domains which were ambulance service, doctors attitude and technology, nursing care, facilities, cost and medical outcome, auxiliary,patient choice and waiting time. The pilot survery showed that Cronbach alpha of 8 factors were 0.8342, 0.8692, 0.8583, 0.4027, 0.6364, 0.3585 and 0.7332 respectively. The KMO statistic of 0.346 suggested what was inadequate to perform factor analysis. In the formal survey 465 interviews were conducted. For the EPSQ , response rate of questionnaires was 96.2%, valid rate of items were 96.8%. Internal consistency reliability coefficient of questionnaires was 0.9471. The internal consistency reliability of the 8 factors were high with a coefficient of 0.8863, 0.8851, 0.8932, 0.7420, 0.7802, 0.7099, 0.8051 and 0.7173 respectively. The content validity andconstruct validity were supported by factor loading .The structure in factor analysis went well with the structure of the questionnaire. The KMO statistic of 0.941 and Bartlett's test of sphericity with Bartlett's x2=6863.3 (P<0.001) suggested that factor analysis could be performed. The convergent validity and discriminate validity were strongly supported by the correlations within the factors which were stronger than that between the factors. The study fits the data of EPSQ and gets the fitting equation followed by Structural Equation Model (SEM) . The Goodness of Fit Index (GFI) of the model was 0.906,and Adjustive Goodness of Fit Index (AGFI) was 0.840. They proclaimed that the model was good.The construct of EPSQ and the relations of 8 factors were confirmed with SEM. , noticeably, doctor care had strong relations with the other factors.Demographic variables predicted dissatisfaction included young age, very high or low income,paying perquisite,high education.Conclusion: The design of EPSQ was reasonable, and its reliability and validity were satisfied. The EPSQ was also practical and could meet the demant in the hospital management.